Case Study
A Friction-Free Employee Experience for 16,000 Workers
Directing a team of eight to deliver an employee intranet that reduced search volume for critical tools by 60%
Role: Lead Strategist & Experience Director
Client: ConocoPhillips
Industry: Energy
Duration: 9 months
Team: 8 (UX, design, strategy, front-end and back-end development)
The Problem
Over a third of ConocoPhillips' 16,000 employees couldn't access the tools, content, or information they needed to do their jobs.
Whether in the office or out in the field, employees needed two things from their intranet: fast access to day-to-day work tools and awareness of company happenings. The existing intranet failed at both. Once cutting-edge, it had aged into something difficult to use, missing key features, stuck on legacy technology, and completely lacking mobile capability or responsive design. Field workers — a massive segment of the workforce — were effectively locked out.
That alone would have been a straightforward redesign. What made it hard was that 15 separate locations each had their own vision for what the intranet should be. Corporate in Houston wanted one thing. Refinery sites wanted another. International offices had different requirements entirely. There was no shared definition of success, no shared understanding of priorities, and no mechanism for getting there.
The design question was clear: how do you build an employee experience platform that works for a globally distributed workforce when 15 different locations can't agree on what "works" means?
My Approach
I started by getting everyone in the same room — or as close to it as possible. I facilitated a two-day workshop that brought together Communications, IT, and HR team members from locations around the world. The goal wasn't to design the intranet in the workshop. It was to surface the actual requirements behind each location's wish list and find the structural overlaps no one had articulated.
That workshop revealed something important: the 15 locations didn't actually disagree on what they needed. They disagreed on what they called it. Once I mapped their requirements against user tasks rather than feature names, the shared needs became obvious — and the location-specific needs became a personalization problem, not a platform problem.
From there, I led user interviews, a company-wide survey, and task analysis to build personas grounded in behavioral patterns across roles and locations. I mapped user journeys to identify where friction lived in the daily workflow and used those journeys to define the information architecture and personalization strategy.
I made a deliberate decision to take a hyper-iterative approach with frequent testing and feedback loops. This wasn't a project where we could afford to reveal a finished product and hope for buy-in. We needed to build confidence incrementally — showing progress, gathering reactions, and adjusting — so that when the platform launched, it already felt like something employees and publishers had co-created. Clickable prototypes went through usability testing cycles, and each round sharpened both the design and the stakeholder confidence.
Key Decisions
Personalization as the architecture, not a feature. The core insight from the workshop was that one intranet couldn't serve 15 locations with a single, static experience. I developed a personalization strategy that tailored news, events, alerts, links, tools, and publications based on each employee's location. This turned the 15-location disagreement from a political problem into a technical one — every location got an intranet that felt like theirs, built on shared infrastructure.
The productivity toolbar. Task analysis showed that employees spent a disproportionate amount of time searching for the same handful of tools and links they used every day. I designed a personalized "My Links" toolbar that surfaced each employee's most critical tools immediately — no hunting, no searching, no navigating three levels of menus. This single decision drove the 60% reduction in search volume. It sounds simple. It was. The hard part was proving through research that this was the highest-impact intervention rather than something more visible and complex.
Self-healing dynamic layouts. I specified an interface pattern where time-sensitive content — alerts, required actions, urgent communications — appeared prominently when relevant and disappeared when resolved, with the layout automatically adjusting to fill the space. This solved two problems at once: important information couldn't be missed, and the interface never felt cluttered with stale content. The component library I directed made this possible — content authors could build and launch pages using a flexible toolbox of responsive components without developer support.
Systemic governance from day one. I implemented a governance model alongside the design system — not after launch. This defined how content was managed, who published what, and how the intranet stayed organized as content grew. Most intranet redesigns look great on launch day and degrade within six months because nobody planned for ongoing content management. I built that plan into the deliverable.
Outcomes
The project delivered an intranet homepage and an employee experience platform that publishers used to build out content and information sites across all 15 locations.
60% reduction in search volume for tools and links housed in the productivity toolbar
Access-anywhere capability with mobile-first responsive design across all devices
Personalized experience for news, events, alerts, links, tools, and publications by location
Overwhelmingly positive response and immediate buy-in from both publishers and content authors
Named to the Intranet Design Annual 2021 — the Year's 10 Best Intranets
The Intranet Design Annual review noted that the new designs clearly addressed the shortcomings of the legacy system, and that the design team was surprised by how immediately and positively users responded — buy-in was instant.
Reflection
This project reinforced something I've seen on every multi-stakeholder engagement: disagreement is usually a vocabulary problem, not a strategy problem. The 15 locations weren't fighting about what the intranet should do — they were using different words to describe the same needs. The two-day workshop didn't resolve the conflict through compromise. It resolved it by reframing the requirements in shared language. Once everyone saw their needs reflected in the same model, alignment happened fast. If I ran this again, I'd embed analytics more aggressively from launch — the 60% search reduction was measured, but I'd want deeper behavioral data on how personalization was performing across locations to feed a continuous optimization loop.